Some contemporary computer-based systems provide for some form of biometric identification. For example, some laptop computers and mobile telephones allow a user to sign in with a fingerprint. If the user's fingerprint is not recognized, the user can try again. This is a one-time sign in, after which all actions are attributed to the authenticated user.
In a dynamic multi-user system with biometric identification, such as a contemporary gaming console or digital advertising billboard, users are biometrically identified (via a face scan for example) during their interactions with the system. After a user is biometrically identified, that user's actions, such as launching an application or placing a phone call, are made on the identified user's behalf.
However, in a dynamic multi-user system, misrecognition in which a user is determined to be someone else is a possibility. As any actions thereafter are specific to a user, incorrect recognition may have significant consequences. For example, a user wants a phone call (e.g., via Skype®) to correctly identify him or her as the caller, wants a product purchased online correctly delivered to his or her address and not someone else's address, wants his or her applications and related data available only to him or her and not someone else, and so on. Even if correctly identified at first, a user is not necessarily identified only once, because for example the user may move in and out of the camera's view; the user may be re-identified in a generally continuous and regularly occurring identification process, providing more chances for misrecognition.